Often times it is desirable to detect changes between video frames in a video sequence. For example, shot-change detection can be implemented to detect changes in video scenes of a video sequence (e.g., to determine a sequence of video frames that are related). In one example, shot-change information can be employed to facilitate distributed transcoding (e.g., splitting a video sequence into chunks of video content and transcoding the chunks of video content in parallel) since it is often times advantageous to split a video sequence for distributed transcoding at shot-change boundaries. However, conventional shot-change detection techniques are computationally demanding. For example, conventional shot-change detection techniques process a video sequence by decoding the video sequence and/or scanning all pixel data in the video sequence. Furthermore, conventional shot-change detection techniques often times compute a pixel luminance histogram for each video frame in the video sequence.